Measures of association which meet Costners P-R-E criterion can ...

[Mimi]

Measures of association which meet Costners P-R-E criterion can be interpreted in a certain way. How are they interpreted?
 
  What will be an ideal response?

Question 2

Suppose the significance level is .02 for the relationship between political party preference and region. What does this mean?
 
  What will be an ideal response?

[vseab]

Answer to Question 1

Answer: Measures of association which meet Costners P-R-E criterion can be interpreted as the proportion by which you can reduce the number of errors in predicting the dependent variable by knowing the independent variable. Thus, if a P-R-E measure of association is .50 for the relationship between support for racial equality and race, this means that we can reduce the number of errors in the dependent variable (support for racial equality) by 50 by knowing the independent variable (race).

Answer to Question 2

Answer: It means that there are 2 chances out of 100 that the relationship between political party preference and region occurred just by chance in the sample and does not really exist in the population from which the sample was selected. Since we want this probability to be .05 or less in order to say that the relationship is statistically significant, this is a statistically significant relationship: we are deciding that there is a real relationship between these two variables in the population from which the sample was selected.

[Mimi]

Thanks for the timely response, appreciate it

[gcook]

Excellent